Comments (3)
First of all, you need to understand the training and test splits used in the field of deep metric learning (DML).
Unlike classification where the images of test set is consist of "seen classes",
The models of DML is evaluated on the task of image retrieval for the "unseen classes".
Therefore, train/test split of CUB200 for DML is different with that of CUB200 for classification.
First 98 classes are used for training, and the last 98 classes are used for test.
You may read the experiments section of the following paper for the details.
https://cvgl.stanford.edu/papers/song_cvpr16.pdf
from rkd.
I see. I'm still a bit confused, however. You said that the first 98 classes are for training while the last 98 are for testing, but what about the remaining 4 classes, are they completely ignored? Because the number of test and train images shown in the code output sum up to exactly how many images there are in the dataset. Sorry if this questions seems dumb.
from rkd.
Sorry for my typo,
I was confused CUB200 with Cars-196
For CUB 200, among 200 classes first 100 classes are used for training, and the last 100 classes are used for test.
For Cars 196, among 196 classes first 98 classes are used for training, and the last 98 classes are used for test.
from rkd.
Related Issues (20)
- Cifar 100 and Tiny Imagenet Training HOT 3
- Hi, there is a similar idea in <<Correlation Congruence for Knowledge Distillation arXiv:1904.01802v1>>. Do you think which is more efficent ? HOT 1
- Few-shot learning experiments
- A question about recall HOT 2
- in loss design , is the code " torch.no_grad " essential ? HOT 1
- EOFError: Ran out of Input
- Student accuracy way below expected HOT 12
- Eval vs Train HOT 1
- What are the numbers in the student and teacher tensors?
- the calculation of relation
- What is its value equal to opts.dark_ratio
- dataset link don't work
- metric/utils.py HOT 2
- can't run the "run_distill.py" HOT 3
- Hi,could you release the cifar-100 classification experiment? HOT 2
- Do you have any suggestions on how to choice the loss ratio of different distill methods? HOT 4
- torch type may not matching in pytorch==1.2 HOT 1
- Can you release your trained model on cub200 ? HOT 1
- A question about paper setting HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from rkd.